Non-small cell lung cancer (NSCLC) is the most frequently-encountered lung cancer, accounting for up to 80-85 percent of all cases. This larger disease is further divided into different subtypes, based on the histology of the cells involved. The most common variants encountered are adenocarcinoma, squamous cell carcinoma, and large cell carcinoma. When one considers all stages of NSCLC at diagnosis, the 5-year survival rate is roughly 15-17 percent.
The presence of distant metastasis has an overall negative impact upon patient outcomes, lowering the 5-year survival rate to roughly 4 percent. One common site of metastasis for NSCLC patients is the brain. The standard first-line therapy for brain metastases has been whole brain radiation therapy, as that technique was shown to effectively manage the neurological symptoms in more than 75 percent of those patients.
Improved survival results have been noted for patients having both their synchronous brain tumor and primary lung tumor lesions resected. However, it is not known what effect the use of adjuvant chemotherapy has on the outcomes of this patient subpopulation.
To address this question, Grace Dy, MD, Division Chief, Thoracic Oncology, Roswell Park Cancer Institute, Buffalo, N.Y., and colleagues conducted a retrospective study using National Cancer Database (NCDB) records from 2010 to 2014. “Our main goal was to obtain a better understanding of the effect that adjuvant chemotherapy had upon the survival of NSCLC patients with synchronous brain metastasis who had their primary lung tumor resected,” Dy explained. The results were presented at the 2017 ASCO Annual Meeting (Abstract 8525).
NSCLC patients who had undergone resection of their primary lung cancer from 2010 to 2014 were identified from the NCDB. When patients with other primary malignancies were excluded, those queried yielded 942,374 individuals. Further limiting these records to those who had a concurrent diagnosis of brain metastasis reduced the number to 46,120 patients. Brain metastasis data in the NCDB was only available from 2010 onwards, thus that was a natural initiation point for this retrospective study. The analysis was further limited to the 751 patients having undergone surgery who were alive at 90 days to eliminate bias from imbalance in early mortality in either arm. These were then limited to those having pathologically-confirmed N stage 0 (n = 303) or stage I (n = 117) disease for a final analysis set of 420 patients.
Treatment group associations were analyzed using either the Wilcoxon rank sum-test for continuous variables (i.e., ones that have an infinite number of possible values) or the Chi-square test for categorical variables (i.e., ones that have a few discrete possible values). The effects of treatment and confounding variables on overall survival (OS) were assessed using univariate and multivariate proportional hazards modeling results. Estimates and 95 percent confidence interval (CI) values for the hazard ratios (HR) were used to summarize the relative prognosis. Kaplan-Meier methodology was utilized for the depiction of unadjusted differences in OS. Results were obtained using the SAS version 9.4 analytical software platform.
Of the 420 patients included in this retrospective study, 292 (69.5%) had received adjuvant chemotherapy, while 128 (30.5%) had not. The mean age for those receiving chemotherapy was 57.7 years, 61.5 years for those who did not, and 58.8 years for the overall group. In the chemotherapy group, 148 (50.7%) of the patients were male and 144 (49.3%) were female; while in the non-chemotherapy group, 67 (52.3%) were male and 61 (47.7%) were female.
Only patients with N stage 0 or I NSCLC were included in these data. For the chemotherapy group, 205 (70.2%) were N stage 0 and 87 (29.8%) were N stage I; for the non-chemotherapy group, 98 (76.6%) were N stage 0 and 30 (23.4%) had N stage I disease. When asked why only N stage 0 or I pathologies were included, Dy responded, “Patients who have N2 disease are typically not surgically resected. If they do in this context, chemotherapy is an accepted standard therapy. We embarked on this analysis as the role of chemotherapy in the N-0 and N-I NSCLC patient population in the context of synchronous brain metastases was not known, because there was no RCT evidence available.”
In the chemotherapy group, 168 patients (57.5%) had open thoracic surgery, 62 (21.2%) had video-assisted thoracoscopic surgery (VATS), and 62 (21.2%) had an unknown type of surgery. The types of surgery were as follows for the non-chemotherapy patients: VATS–35 (27.3%); open surgery–65 (50.8%); unknown type of surgery–28(17.9%). Regarding the two differing surgical approaches, Dy observed: “VATS, in contrast to open approach, has the advantage of smaller surgical incisions, generally quicker post-op recovery times, and reduced risk of thoracotomy pain.”
In the group of patients who received adjuvant chemotherapy, 79 (27.1%) had histologic grade 1/2 disease, 165 (56.5%) had grade 3/4 disease, and 48 (16.4%) had disease of unknown grade. For the non-chemotherapy patients, 39 (30.5%) had grade 1/2 disease, 77 (60.2%) had grade 3/4 disease, and 12 (9.4%) had disease of unknown grade.
The 1-year survival rate was 84 percent (95% CI – 79-88%) for the chemotherapy group, 68 percent (95% CI – 57-77%) for the non-chemotherapy group, and 80 percent (95% CI – 75-84%) for the entire group. The 5-year survival rate for the chemotherapy group was 34 percent (95% CI – 26-42%), while for the non-chemotherapy group, the figure was 27 percent (95% CI – 17-38%).
Similar to the 1- and 5-year survival rates, the median OS was higher for patients receiving adjuvant chemotherapy. These patients had a median OS of 34.4 months (95% CI – 26.3-41.0 months), while those not receiving chemotherapy had a median OS of 28.1 months (95% CI – 14.3-28.1 months).
Univariate analysis was performed to determine which variables had an impact on the OS of the patients in this retrospective study. In this analysis set, there was a very clear improvement in OS for those having chemotherapy, with a HR of 0.61 (reference group: no chemotherapy; 95% CI – 0.45-0.84; P = 0.002). The histology of the patient also appeared to have an effect on their OS, with a HR of 0.55 favoring adenocarcinoma histology (reference group: squamous histology; 95% CI – 0.37-0.84; p = 0.005). In this analysis set, the N pathology grade had an impact, favoring patients with N stage 0 over those having N stage I pathology, with a HR of 1.44 (reference group: N stage 0; 95% CI – 1.05-1.98; p = 0.025). There was also a slight advantage for younger patients in the univariate analysis, with a HR of 1.03 (95% CI – 1.02-1.05; p < 0.001).
Multivariate analysis was also performed to evaluate the effects of certain variables upon the patients' OS. As in the aforementioned univariate analysis, there was a clear benefit for those who underwent chemotherapy, with an HR of 0.60 (reference: no chemotherapy; 95% CI – 0.43-0.83; P = 0.002). In this analysis set, the histology of the disease also had an impact on OS, favoring those with adenocarcinoma histology (reference: squamous histology; 95% CI – 0.40-0.92; p = 0.019). This analysis also showed that age had a slight impact on the OS of the patients, with an HR of 1.03 (95% CI – 1.01-1.05; p < 0.001).
When asked to comment on the univariate and multivariate analyses results, Dy noted, “Univariate analyses revealed that age, pathologic N status, histologic type, and administration of chemotherapy ‘adjuvantly’ in this setting were correlated with OS; multivariate analysis showed age, histologic type, and administration of chemotherapy adjuvantly were correlated with OS.”
Dy had the following observation regarding the effect of NSCLC histology on OS: “Prognostically, the squamous histologic type has a known association with worse survival in various clinical trials. A hypothesis for why this could be is that squamous histology generally is diagnosed in active smokers who often have more comorbid health issues relative to those patients with adenocarcinoma, where there are subsets of patients who are never-smokers or former smokers.
“The central location of squamous histology also poses greater risk for clinical complications such as obstruction, hemoptysis/bleeding,” she further added.
Those patients receiving adjuvant chemotherapy had better 1-year and 5-year survival rates (84% and 34%) than their counterparts who did not receive chemotherapy (68%, 27%). Generally speaking, those in the chemotherapy group tended to be younger than those in the non-chemotherapy group (median 58 years vs. 62 years, p < 0.001). Nonetheless, chemotherapy administration remained an independent factor associated with survival benefit. When asked about the types of chemotherapy used, Dy replied, “This study did not distinguish between the chemotherapy treatments used, as the NCDB does not have granularity about regimen choice.
“There is obviously inherent bias with this type of analysis; however, we were reassured to find that there was no apparent detrimental effect for ‘adjuvant’ chemotherapy in this high-risk group of patients. [This] is particularly relevant since in the conventional adjuvant systemic therapy setting—in patients without distant metastasis—there is potential harm when adjuvant chemotherapy is administered to subsets of patients with pathologic N0 status—that is, stage IA disease T1N0.”
Richard Simoneaux is a contributing writer.